English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

We present MRGAN, a multi-rooted adversarial network which generatespart-disentangled 3D point-cloud shapes without part-based shape supervision.The network fuses multiple branches of tree-structured graph convolution layerswhich produce point clouds, with learnable constant inputs at the tree roots.Each branch learns to grow a different shape part, offering control over theshape generation at the part level. Our network encourages disentangledgeneration of semantic parts via two key ingredients: a root-mixing trainingstrategy which helps decorrelate the different branches to facilitatedisentanglement, and a set of loss terms designed with part disentanglement andshape semantics in mind. Of these, a novel convexity loss incentivizes thegeneration of parts that are more convex, as semantic parts tend to be. Inaddition, a root-dropping loss further ensures that each root seeds a singlepart, preventing the degeneration or over-growth of the point-producingbranches. We evaluate the performance of our network on a number of 3D shapeclasses, and offer qualitative and quantitative comparisons to previous worksand baseline approaches. We demonstrate the controllability offered by ourpart-disentangled generation through two applications for shape modeling: partmixing and individual part variation, without receiving segmented shapes asinput.

MRGAN: Multi-Rooted 3D Shape Generation with Unsupervised Part Disentanglement